4 Steps to Creating Effective BI TeamsJuly 15, 2012 by Michael Koploy
Business intelligence–the use of sophisticated software to analyze complex data–is no longer the domain of a centralized group of IT staff or advanced data analysts. Today, powerful and Web-based BI tools are accessible to a wide range of business users.
BI is everywhere, and it’s everyone’s job. But with this proliferation comes new challenges. Teams of BI users today often lack the structure, guidance and leadership to effectively mine data. In this article, I’ll share four steps to establish guidelines, organize teams, delegate data management and allow the success of the BI team to permeate and drive innovation throughout the business.
1. Establish Structure and Boundaries with Governance
Structure is necessary both to establish research boundaries and accomplish BI project objectives. Defining important roles and responsibilities is a critical first step that is often not given enough attention, according to Myron Weber, founder of BI advisory firm Northwood Advisors.
“It’s absolutely essential that stakeholders know what they can and cannot do, what they should and should not do,” says Weber. However, he notes the best way to address governance differs and often depends on the existing management structure. Larger organizations may require a strict set of written guidelines for employees and a chain of command that determines who can initiate a project, and when.
For smaller businesses, the right structure may simply be face-to-face meetings. David Handmaker, CEO of Next Day Flyers, holds weekly meetings with his analysts to review project status and discuss near- and long-term strategies to address any issues.
2. Match Technical Resources to Domain Experts
Individuals in sales, marketing and operations are increasingly using BI applications to improve their group’s effectiveness. While these individuals generally possess a deeper understanding of their role in the business than IT, they frequently lack experience using BI tools.
That’s why companies should pair business users with technical resources. These resources can assist users with questions about, for instance, applying the appropriate statistical model or pulling the correct data set. Whether you should use internal experts or outside resources (typically consultants or your BI vendor) is largely a matter of preference and your staff's proficiencies.
At Next Day Flyers, Handmaker relies on the technical expertise of a sole senior business intelligence analyst to assist his sales and marketing team with the proper use of the MicroStrategy application. If an employee has a question about running a particular regression or needs to create a new custom report, Handmaker instructs them to coordinate with the lead analyst.
Bryan Bayless, VP of GTM Finance at McAfee, chose to rely on the support team of their application provider, Anaplan. Bayless leads a team mostly made-up of financial experts, not traditional BI users, and was charged to improve the software company’s internal commission model. He notes the ease of use of their tool and the help of the Anaplan team allows him to instead fill his team with employees experienced with McAfee’s sales process.
3. Empower Business Users to Own Data Management
Tom Kindermans, SVP of Business Solutions for SAP Asia Pacific and Japan, recently told CIO India that proper data cleansing was one of the biggest enterprise concerns of the next decade. However, Weber points out that too many organizations unnecessarily obsess over data preparation, estimating that many analysts spend 60 to 90 percent of their time preparing for analysis.
“If directionality is all that is necessary for action, why obsess over data accuracy?” he questions, calling the pursuit of “perfect data” a wasteful use of resources.
The solution, says Weber, is to have business users work with the IT team to determine what makes data “correct, acceptable and accurate.”
Handmaker took the responsibility for data management out of IT entirely. He placed one of his business analysts in charge of a nightly extract, transform and load (ETL) pull of its internal data because he felt this individual was better able to catch errors and correctly identify when data becomes “too dirty.”
“The business analyst should direct IT–not the other way around,” says Handmaker.
4. Find New Holes for BI Pegs
Organizations that develop an effective governance framework, designate expert leadership, and effectively prioritize data management will often find their BI teams to be a valuable strategic resource–but one that is often underutilized. In what he describes as “pivoting,” Weber notes that mature BI groups “evolve their charter over time.”
“These groups drive better decision-making processes throughout the enterprise not just by providing better information, but by re-defining the process through which questions are first asked and action is taken,” says Weber.
Perhaps more so than other strategic groups in an organization, BI teams are well-suited to apply their processes to improve company performance in new areas.
Bayless says this is currently happening within his team at McAfee. After achieving success with its initial charter of improving the sales commission process, McAfee’s executive leadership asked the BI team to analyze the territorial coverage of the regional sales teams. Bayless’s team is now delivering a greater return on investment than initially envisioned.
What have you done at your organization to ensure the organization is effectively analyzing its data? If you have any additional thoughts on how to best organize roles and responsibilities within teams of BI users, please share them below.
Thumbnail image created by genvessel.